import pandas as pd, pickle, numpy as np, matplotlib.pyplot as plt, matplotlib.dates as mdates, matplotlib as mpl
import matplotlib.lines as mlines
from datetime import datetime, timedelta

Visual Analysis of the Coronavirus Pandemic (State and County Level)

• CDC cases and death information
• Safegraph county census data
• Google mobility data

Update Data and Import Graphing Functions

#%run safegraph_census.ipynb

#%run create_master_table.ipynb

with open('covid_data.p', 'rb') as f:
    data = pickle.load(f)

%run graphFunctions.ipynb

print("Total Deaths :", data[data.date == data.date.max()- timedelta(days=0)].deaths_cdc.sum())

Total Deaths : 423566.0

Section 1: Statewide Totals

Figure 1.1: Deaths by State

plotStateTotals(data, 'deaths_cdc', 'deaths_cdc', 200)

Figure 1.2: Deaths per 100,000 People

plotStateTotals(data, 'deaths_cdc', 'deaths_cdc_per_100k', 200)

Figure 1.3: Cases and Deaths (Total, Per 100k People) - All States

states=data.state_code.unique()
stateGraphs(data, states,'confirmed_cdc','deaths_cdc', '2020-03-01', 21)

Figure 1.4: Cases and Deaths (Total, Per 100k People) - Northeast

states = getStatesInRegion("N").keys()
stateGraphs(data, states,'confirmed_cdc','deaths_cdc', '2020-03-01', 21)

Figure 1.5: Cases and Deaths (Total, Per 100k People) - South

states = getStatesInRegion("S").keys()
stateGraphs(data, states,'confirmed_cdc','deaths_cdc', '2020-03-01', 21)

Figure 1.6: Cases and Deaths (Total, Per 100k People) - Midwest

states = getStatesInRegion("M").keys()
stateGraphs(data, states,'confirmed_cdc','deaths_cdc', '2020-03-01', 21)

Figure 1.7: Cases and Deaths (Total, Per 100k People) - West

states = getStatesInRegion("W").keys()
stateGraphs(data, states,'confirmed_cdc','deaths_cdc', '2020-03-01', 21)

Figure 1.8 Cases and Deaths (Total, Per 100k People) - Other

states = getStatesInRegion("O").keys()
stateGraphs(data, states,'confirmed_cdc','deaths_cdc', '2020-03-01', 21)

#

Section 2: Deep Dive on States with Top 10 Death Count

#

#

Section 2.1: New York

state = 'NY'
death_min = 500

Figure 2.1.1: Cases and Deaths (Total, Per 100k People)

stateGraphs(data, [state],'confirmed_cdc','deaths_cdc', '2020-03-01', 7)

Figure 2.1.2: Deaths per 100k People - 14 Day Average

plotCountyDeathCurves(data, state, death_min = death_min, rolling_ave = 14, start_date='2020-03-14')

Section 2.1.3: County Level Mobility Reduction

Mobility Reduction by County - Showing daily fluctuations

graphMobilityCounty(data, state, '2020-03-01', 1, death_min = death_min)

Mobility Reduction by County - 14 Day Average

graphMobilityCounty(data, state, '2020-03-01', 14, death_min = death_min)

Section 2.1.4: County Attributes Analysis

Figure 2.1.4.1: Population Density and Over Age 55 Percentile

• The purple to yellow color gradient is the date, with yellow being the most recent
• The red color gradient is deaths per 100k people

plotInteractions(data, 'pop_density_percentile', 'age_55_plus_pct_percentile', 'deaths_per_100k',[state])

Figure 2.1.4.2: Population Density and (Non) White Percentile

• The purple to yellow color gradient is the date, with yellow being the most recent
• The red color gradient is deaths per 100k people

plotInteractions(data, 'pop_density_percentile', 'r_white_pct_percentile', 'deaths_per_100k',[state])

Figure 2.1.4.3: Population Density and Uninsured Percentile

• The purple to yellow color gradient is the date, with yellow being the most recent
• The red color gradient is deaths per 100k people

plotInteractions(data, 'pop_density_percentile', 'unins_pct_percentile', 'deaths_per_100k',[state])

Figure 2.1.4.4: Uninsured Percentile and Poverty Income Ratio > 200% Percentile

• The purple to yellow color gradient is the date, with yellow being the most recent
• The red color gradient is deaths per 100k people

plotInteractions(data, 'unins_pct_percentile', 'pir_200_plus_pct_percentile', 'deaths_per_100k',[state])

# # # #

Section 2.2: California

state = 'CA'
death_min = 600

Figure 2.2.1: Cases and Deaths (Total, Per 100k People)

stateGraphs(data, [state],'confirmed_cdc','deaths_cdc', '2020-03-01', 7)

Figure 2.2.2: Deaths per 100k People - 14 Day Average

plotCountyDeathCurves(data, state, death_min = death_min, rolling_ave = 14, start_date='2020-03-14')

Section 2.2.3: County Level Mobility Reduction

Mobility Reduction by County - Showing daily fluctuations

graphMobilityCounty(data, state, '2020-03-01', 1, death_min = death_min)

Mobility Reduction by County - 14 Day Average

graphMobilityCounty(data, state, '2020-03-01', 14, death_min = death_min)

Section 2.2.4: County Attributes Analysis

Figure 2.2.4.1: Population Density and Over Age 55 Percentile

• The purple to yellow color gradient is the date, with yellow being the most recent
• The red color gradient is deaths per 100k people

plotInteractions(data, 'pop_density_percentile', 'age_55_plus_pct_percentile', 'deaths_per_100k',[state])

Figure 2.2.4.2: Population Density and (Non) White Percentile

• The purple to yellow color gradient is the date, with yellow being the most recent
• The red color gradient is deaths per 100k people

plotInteractions(data, 'pop_density_percentile', 'r_white_pct_percentile', 'deaths_per_100k',[state])

Figure 2.2.4.3: Population Density and Uninsured Percentile

• The purple to yellow color gradient is the date, with yellow being the most recent
• The red color gradient is deaths per 100k people

plotInteractions(data, 'pop_density_percentile', 'unins_pct_percentile', 'deaths_per_100k',[state])

Figure 2.2.4.4: Uninsured Percentile and Poverty Income Ratio > 200% Percentile

• The purple to yellow color gradient is the date, with yellow being the most recent
• The red color gradient is deaths per 100k people

plotInteractions(data, 'unins_pct_percentile', 'pir_200_plus_pct_percentile', 'deaths_per_100k',[state])

# # # #

Section 2.3: Texas

state = 'TX'
death_min = 500

Figure 2.3.1: Cases and Deaths (Total, Per 100k People)

stateGraphs(data, [state],'confirmed_cdc','deaths_cdc', '2020-03-01', 7)

Figure 2.3.2: Deaths per 100k People - 14 Day Average

plotCountyDeathCurves(data, state, death_min = death_min, rolling_ave = 14, start_date='2020-03-14')

Section 2.3.3: County Level Mobility Reduction

Mobility Reduction by County - Showing daily fluctuations

graphMobilityCounty(data, state, '2020-03-01', 1, death_min = death_min)

Mobility Reduction by County - 14 Day Average

graphMobilityCounty(data, state, '2020-03-01', 14, death_min = death_min)

Section 2.3.4: County Attributes Analysis

Figure 2.3.4.1: Population Density and Over Age 55 Percentile

• The purple to yellow color gradient is the date, with yellow being the most recent
• The red color gradient is deaths per 100k people

plotInteractions(data, 'pop_density_percentile', 'age_55_plus_pct_percentile', 'deaths_per_100k',[state])

Figure 2.3.4.2: Population Density and (Non) White Percentile

• The purple to yellow color gradient is the date, with yellow being the most recent
• The red color gradient is deaths per 100k people

plotInteractions(data, 'pop_density_percentile', 'r_white_pct_percentile', 'deaths_per_100k',[state])

Figure 2.3.4.3: Population Density and Uninsured Percentile

• The purple to yellow color gradient is the date, with yellow being the most recent
• The red color gradient is deaths per 100k people

plotInteractions(data, 'pop_density_percentile', 'unins_pct_percentile', 'deaths_per_100k',[state])

Figure 2.3.4.4: Uninsured Percentile and Poverty Income Ratio > 200% Percentile

• The purple to yellow color gradient is the date, with yellow being the most recent
• The red color gradient is deaths per 100k people

plotInteractions(data, 'unins_pct_percentile', 'pir_200_plus_pct_percentile', 'deaths_per_100k',[state])

# # # #

Section 2.4: Florida

state = 'FL'
death_min = 500

Figure 2.4.1: Cases and Deaths (Total, Per 100k People)

stateGraphs(data, [state],'confirmed_cdc','deaths_cdc', '2020-03-01', 7)

Figure 2.4.2: Deaths per 100k People - 14 Day Average

plotCountyDeathCurves(data, state, death_min = death_min, rolling_ave = 14, start_date='2020-03-14')

Section 2.4.3: County Level Mobility Reduction

Mobility Reduction by County - Showing daily fluctuations

graphMobilityCounty(data, state, '2020-03-01', 1, death_min = death_min)

Mobility Reduction by County - 14 Day Average

graphMobilityCounty(data, state, '2020-03-01', 14, death_min = death_min)

Section 2.4.4: County Attributes Analysis

Figure 2.4.4.1: Population Density and Over Age 55 Percentile

• The purple to yellow color gradient is the date, with yellow being the most recent
• The red color gradient is deaths per 100k people

plotInteractions(data, 'pop_density_percentile', 'age_55_plus_pct_percentile', 'deaths_per_100k',[state])

Figure 2.4.4.2: Population Density and (Non) White Percentile

• The purple to yellow color gradient is the date, with yellow being the most recent
• The red color gradient is deaths per 100k people

plotInteractions(data, 'pop_density_percentile', 'r_white_pct_percentile', 'deaths_per_100k',[state])

Figure 2.4.4.3: Population Density and Uninsured Percentile

• The purple to yellow color gradient is the date, with yellow being the most recent
• The red color gradient is deaths per 100k people

plotInteractions(data, 'pop_density_percentile', 'unins_pct_percentile', 'deaths_per_100k',[state])

Figure 2.4.4.4: Uninsured Percentile and Poverty Income Ratio > 200% Percentile

• The purple to yellow color gradient is the date, with yellow being the most recent
• The red color gradient is deaths per 100k people

plotInteractions(data, 'unins_pct_percentile', 'pir_200_plus_pct_percentile', 'deaths_per_100k',[state])

# # # #

Section 2.5: New Jersey

state = 'NJ'
death_min = 500

Figure 2.5.1: Cases and Deaths (Total, Per 100k People)

stateGraphs(data, [state],'confirmed_cdc','deaths_cdc', '2020-03-01', 7)

Figure 2.5.2: Deaths per 100k People - 14 Day Average

plotCountyDeathCurves(data, state, death_min = death_min, rolling_ave = 14, start_date='2020-03-14')

Section 2.5.3: County Level Mobility Reduction

Mobility Reduction by County - Showing daily fluctuations

graphMobilityCounty(data, state, '2020-03-01', 1, death_min = death_min)

Mobility Reduction by County - 14 Day Average

graphMobilityCounty(data, state, '2020-03-01', 14, death_min = death_min)

Section 2.5.4: County Attributes Analysis

Figure 2.5.4.1: Population Density and Over Age 55 Percentile

• The purple to yellow color gradient is the date, with yellow being the most recent
• The red color gradient is deaths per 100k people

plotInteractions(data, 'pop_density_percentile', 'age_55_plus_pct_percentile', 'deaths_per_100k',[state])

Figure 2.5.4.2: Population Density and (Non) White Percentile

• The purple to yellow color gradient is the date, with yellow being the most recent
• The red color gradient is deaths per 100k people

plotInteractions(data, 'pop_density_percentile', 'r_white_pct_percentile', 'deaths_per_100k',[state])

Figure 2.5.4.3: Population Density and Uninsured Percentile

• The purple to yellow color gradient is the date, with yellow being the most recent
• The red color gradient is deaths per 100k people

plotInteractions(data, 'pop_density_percentile', 'unins_pct_percentile', 'deaths_per_100k',[state])

Figure 2.5.4.4: Uninsured Percentile and Poverty Income Ratio > 200% Percentile

• The purple to yellow color gradient is the date, with yellow being the most recent
• The red color gradient is deaths per 100k people

plotInteractions(data, 'unins_pct_percentile', 'pir_200_plus_pct_percentile', 'deaths_per_100k',[state])

# # # #

Section 2.6: Pennsylvania

state = 'PA'
death_min = 500

Figure 2.6.1: Cases and Deaths (Total, Per 100k People)

stateGraphs(data, [state],'confirmed_cdc','deaths_cdc', '2020-03-01', 7)

Figure 2.6.2: Deaths per 100k People - 14 Day Average

plotCountyDeathCurves(data, state, death_min = death_min, rolling_ave = 14, start_date='2020-03-14')

Section 2.6.3: County Level Mobility Reduction

Mobility Reduction by County - Showing daily fluctuations

graphMobilityCounty(data, state, '2020-03-01', 1, death_min = death_min)

Mobility Reduction by County - 14 Day Average

graphMobilityCounty(data, state, '2020-03-01', 14, death_min = death_min)

Section 2.6.4: County Attributes Analysis

Figure 2.6.4.1: Population Density and Over Age 55 Percentile

• The purple to yellow color gradient is the date, with yellow being the most recent
• The red color gradient is deaths per 100k people

plotInteractions(data, 'pop_density_percentile', 'age_55_plus_pct_percentile', 'deaths_per_100k',[state])

Figure 2.6.4.2: Population Density and (Non) White Percentile

• The purple to yellow color gradient is the date, with yellow being the most recent
• The red color gradient is deaths per 100k people

plotInteractions(data, 'pop_density_percentile', 'r_white_pct_percentile', 'deaths_per_100k',[state])

Figure 2.6.4.3: Population Density and Uninsured Percentile

• The purple to yellow color gradient is the date, with yellow being the most recent
• The red color gradient is deaths per 100k people

plotInteractions(data, 'pop_density_percentile', 'unins_pct_percentile', 'deaths_per_100k',[state])

Figure 2.6.4.4: Uninsured Percentile and Poverty Income Ratio > 200% Percentile

• The purple to yellow color gradient is the date, with yellow being the most recent
• The red color gradient is deaths per 100k people

plotInteractions(data, 'unins_pct_percentile', 'pir_200_plus_pct_percentile', 'deaths_per_100k',[state])

# # # #

Section 2.7: Illinois

state = 'IL'
death_min = 300

Figure 2.7.1: Cases and Deaths (Total, Per 100k People)

stateGraphs(data, [state],'confirmed_cdc','deaths_cdc', '2020-03-01', 7)

Figure 2.7.2: Deaths per 100k People - 14 Day Average

plotCountyDeathCurves(data, state, death_min = death_min, rolling_ave = 14, start_date='2020-03-14')

Section 2.7.3: County Level Mobility Reduction

Mobility Reduction by County - Showing daily fluctuations

graphMobilityCounty(data, state, '2020-03-01', 1, death_min = death_min)

Mobility Reduction by County - 14 Day Average

graphMobilityCounty(data, state, '2020-03-01', 14, death_min = death_min)

Section 2.7.4: County Attributes Analysis

Figure 2.7.4.1: Population Density and Over Age 55 Percentile

• The purple to yellow color gradient is the date, with yellow being the most recent
• The red color gradient is deaths per 100k people

plotInteractions(data, 'pop_density_percentile', 'age_55_plus_pct_percentile', 'deaths_per_100k',[state])

Figure 2.7.4.2: Population Density and (Non) White Percentile

• The purple to yellow color gradient is the date, with yellow being the most recent
• The red color gradient is deaths per 100k people

plotInteractions(data, 'pop_density_percentile', 'r_white_pct_percentile', 'deaths_per_100k',[state])

Figure 2.7.4.3: Population Density and Uninsured Percentile

• The purple to yellow color gradient is the date, with yellow being the most recent
• The red color gradient is deaths per 100k people

plotInteractions(data, 'pop_density_percentile', 'unins_pct_percentile', 'deaths_per_100k',[state])

Figure 2.7.4.4: Uninsured Percentile and Poverty Income Ratio > 200% Percentile

• The purple to yellow color gradient is the date, with yellow being the most recent
• The red color gradient is deaths per 100k people

plotInteractions(data, 'unins_pct_percentile', 'pir_200_plus_pct_percentile', 'deaths_per_100k',[state])

# # # #

Section 2.8: Massachusetts

state = 'MA'
death_min = 300

Figure 2.8.1: Cases and Deaths (Total, Per 100k People)

stateGraphs(data, [state],'confirmed_cdc','deaths_cdc', '2020-03-01', 7)

Figure 2.8.2: Deaths per 100k People - 14 Day Average

plotCountyDeathCurves(data, state, death_min = death_min, rolling_ave = 14, start_date='2020-03-14')

Section 2.8.3: County Level Mobility Reduction

Mobility Reduction by County - Showing daily fluctuations

graphMobilityCounty(data, state, '2020-03-01', 1, death_min = death_min)

Mobility Reduction by County - 14 Day Average

graphMobilityCounty(data, state, '2020-03-01', 14, death_min = death_min)

Section 2.8.4: County Attributes Analysis

Figure 2.8.4.1: Population Density and Over Age 55 Percentile

• The purple to yellow color gradient is the date, with yellow being the most recent
• The red color gradient is deaths per 100k people

plotInteractions(data, 'pop_density_percentile', 'age_55_plus_pct_percentile', 'deaths_per_100k',[state])

Figure 2.8.4.2: Population Density and (Non) White Percentile

• The purple to yellow color gradient is the date, with yellow being the most recent
• The red color gradient is deaths per 100k people

plotInteractions(data, 'pop_density_percentile', 'r_white_pct_percentile', 'deaths_per_100k',[state])

Figure 2.8.4.3: Population Density and Uninsured Percentile

• The purple to yellow color gradient is the date, with yellow being the most recent
• The red color gradient is deaths per 100k people

plotInteractions(data, 'pop_density_percentile', 'unins_pct_percentile', 'deaths_per_100k',[state])

Figure 2.8.4.4: Uninsured Percentile and Poverty Income Ratio > 200% Percentile

• The purple to yellow color gradient is the date, with yellow being the most recent
• The red color gradient is deaths per 100k people

plotInteractions(data, 'unins_pct_percentile', 'pir_200_plus_pct_percentile', 'deaths_per_100k',[state])

# # # #

Section 2.9: Arizona

state = 'AZ'
death_min = 300

Figure 2.9.1: Cases and Deaths (Total, Per 100k People)

stateGraphs(data, [state],'confirmed_cdc','deaths_cdc', '2020-03-01', 7)

Figure 2.9.2: Deaths per 100k People - 14 Day Average

plotCountyDeathCurves(data, state, death_min = death_min, rolling_ave = 14, start_date='2020-03-14')

Section 2.9.3: County Level Mobility Reduction

Mobility Reduction by County - Showing daily fluctuations

graphMobilityCounty(data, state, '2020-03-01', 1, death_min = death_min)

Mobility Reduction by County - 14 Day Average

graphMobilityCounty(data, state, '2020-03-01', 14, death_min = death_min)

Section 2.9.4: County Attributes Analysis

Figure 2.1.4.1: Population Density and Over Age 55 Percentile

• The purple to yellow color gradient is the date, with yellow being the most recent
• The red color gradient is deaths per 100k people

plotInteractions(data, 'pop_density_percentile', 'age_55_plus_pct_percentile', 'deaths_per_100k',[state])

Figure 2.9.4.2: Population Density and (Non) White Percentile

• The purple to yellow color gradient is the date, with yellow being the most recent
• The red color gradient is deaths per 100k people

plotInteractions(data, 'pop_density_percentile', 'r_white_pct_percentile', 'deaths_per_100k',[state])

Figure 2.9.4.3: Population Density and Uninsured Percentile

• The purple to yellow color gradient is the date, with yellow being the most recent
• The red color gradient is deaths per 100k people

plotInteractions(data, 'pop_density_percentile', 'unins_pct_percentile', 'deaths_per_100k',[state])

Figure 2.9.4.4: Uninsured Percentile and Poverty Income Ratio > 200% Percentile

• The purple to yellow color gradient is the date, with yellow being the most recent
• The red color gradient is deaths per 100k people

plotInteractions(data, 'unins_pct_percentile', 'pir_200_plus_pct_percentile', 'deaths_per_100k',[state])

# # # #

Section 2.10: Michigan

state = 'MI'
death_min = 300

Figure 2.10.1: Cases and Deaths (Total, Per 100k People)

stateGraphs(data, [state],'confirmed_cdc','deaths_cdc', '2020-03-01', 7)

Figure 2.10.2: Deaths per 100k People - 14 Day Average

plotCountyDeathCurves(data, state, death_min = death_min, rolling_ave = 14, start_date='2020-03-14')

Section 2.10.3: County Level Mobility Reduction

Mobility Reduction by County - Showing daily fluctuations

graphMobilityCounty(data, state, '2020-03-01', 1, death_min = death_min)

Mobility Reduction by County - 14 Day Average

graphMobilityCounty(data, state, '2020-03-01', 14, death_min = death_min)

Section 2.10.4: County Attributes Analysis

Figure 2.1.4.1: Population Density and Over Age 55 Percentile

• The purple to yellow color gradient is the date, with yellow being the most recent
• The red color gradient is deaths per 100k people

plotInteractions(data, 'pop_density_percentile', 'age_55_plus_pct_percentile', 'deaths_per_100k',[state])

Figure 2.10.4.2: Population Density and (Non) White Percentile

• The purple to yellow color gradient is the date, with yellow being the most recent
• The red color gradient is deaths per 100k people

plotInteractions(data, 'pop_density_percentile', 'r_white_pct_percentile', 'deaths_per_100k',[state])

Figure 2.10.4.3: Population Density and Uninsured Percentile

• The purple to yellow color gradient is the date, with yellow being the most recent
• The red color gradient is deaths per 100k people

plotInteractions(data, 'pop_density_percentile', 'unins_pct_percentile', 'deaths_per_100k',[state])

Figure 2.10.4.4: Uninsured Percentile and Poverty Income Ratio > 200% Percentile

• The purple to yellow color gradient is the date, with yellow being the most recent
• The red color gradient is deaths per 100k people

plotInteractions(data, 'unins_pct_percentile', 'pir_200_plus_pct_percentile', 'deaths_per_100k',[state])

# # # #

Section 3: State Caparison by Region (Northeast, South, Midwest, West, Other)

Figure 3.1: Deaths - 14 Day Average - Northeast

states = getStatesInRegion("N").keys()
statesCompare(data, states, 'confirmed_cdc', 'deaths_cdc', '2020-03-01', 14)

Figure 3.2: Deaths - 14 Day Average - South

states = getStatesInRegion("S").keys()
statesCompare(data, states, 'confirmed_cdc', 'deaths_cdc', '2020-03-01', 14)

Figure 3.3: Deaths - 14 Day Average - Midwest

states = getStatesInRegion("M").keys()
statesCompare(data, states, 'confirmed_cdc', 'deaths_cdc', '2020-03-01', 14)

Figure 3.4: Deaths - 14 Day Average - West

states = getStatesInRegion("W").keys()
statesCompare(data, states, 'confirmed_cdc', 'deaths_cdc', '2020-03-01', 14)

Figure 3.5: Deaths - 14 Day Average - Other

states = getStatesInRegion("O").keys()
statesCompare(data, states, 'confirmed_cdc', 'deaths_cdc', '2020-03-01', 14)

# # # #

Section 4: County Attribute Analysis Across All States

• Available fields: pir_grp, unins_grp, pop_density_grp, e_grp, r_b_grp, r_w_grp, age_55_plus_grp

states = list(data.state_code.unique())

Figure 4.1: Population Density

plotGroupedDeathCurves(data, states=states, rolling_ave=30, bycol='pop_density_grp', start_date='2020-03-01')

Figure 4.2: Race

The race definition used here is 'white' as reported in census data. Low would be 'non-white' and high would be 'white'.

plotGroupedDeathCurves(data, states=states, rolling_ave=30, bycol='r_w_grp', start_date='2020-03-01')

Figure 4.2: Uninsured

'Low' are the counties with the lowest percentages of uninsured individuals.

plotGroupedDeathCurves(data, states=states, rolling_ave=30, bycol='unins_grp', start_date='2020-03-01')

Figure 4.3: Poverty Income Ratio

'High' are the counties with the highest percentage of PIR > 200%

plotGroupedDeathCurves(data, states=states, rolling_ave=30, bycol='pir_grp', start_date='2020-03-01')

Figure 4.4: Age

plotGroupedDeathCurves(data, states=states, rolling_ave=30, bycol='age_55_plus_grp', start_date='2020-03-01')